At 02:16 AM 06/07/2001 -0700, you wrote:
>So, since it's on my mind again: how fast and in what sized portions can you
>replace parts of your living brain with artificial neurons and not have
>discovered an extremely novel and expensive way of killing yourself? Take
>out half the brain, destructively copy it, put the copy back in -- sounds
>like killing yourself to me. You'll have left a copy behind, but the
>essential you is dead. One neuron per minute sounds reasonable. It's an
>interesting thing to think about -- how much of a system can you replace at
>once with having been said to have destroyed the original system?
Why would you need to destroy the original?
There are scanning techniques that can resolve down to individual neurons
in insects non-destructively now. They require pretty intense magnetic
fields and only work on small things so far, but I imagine scaling it up is
not an insoluble problem. We would also need to scan at a fair bit below
the level of detail of neurons so that you make out receptor sites and
synaptic vesicles. That may be enough to automate recording of a human
brain so that a model of its operation could be accurately be recreated.
This way you could have a copy of yourself made so as insurance against a
calamity. You would periodically "backup" your brain to record all the more
recent memories and thought patterns, and choose at which model contains
the happiest and healthiest version of yourself to be used as your final
self upon biological death.
The model would operate inside a computer (likely not the kind of computer
we have at the moment), and would play inside virtual worlds. There would
be windows out onto the "real" world too.
I like very much the idea that I could keep a backup version of myself
without endangering my biological self in the process. I hope it is not too
far away. I figure maybe 20 years from now...
Q. What is the similarity between an elephant and a grape?
A. They are both purple... except for the elephant.
Virtual Reality Association http://www.vr.org.au
This archive was generated by hypermail 2b30 : Fri Oct 12 2001 - 14:39:42 MDT